Intensity of economic relationships: a spatial econometric analysis of regional economic growth in China

Since inter-regional commodity trade within a country can flow freely, the intensity of economic relationships (IER) between regions may affect the level of the regional convergence. This relationship is explored using an extended spatial Durbin model with 1992-2010 prefecture data in China. Two sample cases are conducted for spatial heterogeneity. We find that while the intensity of economic relationships plays little role in regional growth, the spatial lag of the intensity of economic relationships does matter. Further, the positive contribution of IER to regional growth gradually increases over time. In addition, IER has a positive effect on convergence velocity at the prefectural level but a negative effect at the provincial level. The modifiable areal unit problem is addressed with both scale effects (provincial level data are adopted) and zoning effect (two kinds of delineation of the study area); for the latter effect, a case study of Central China with two samples from the central region and the central core region was analyzed using zoning effect tests. Bayesian comparison analysis results support the finding that the data-driven model results are consistent with the theory-driven model. Our conclusions are robust with respect to alternative fixed effects, measurement of intensity of economic relationships, and the choice of the spatial weight matrix, but vary across types of spatial units due to market segmentation.